Thursday, September 19, 2013

Stages!


I've made some quick updates to the prototype.

Move drafting is finished, there are now six moves that are reasonably balanced. (Reasonably balanced means that every move except the blue Line Shot was called OP by at least one person in testing, and I think every move was called trash at least once as well.)

We have five stages now, which is partly an attempt to stop the disengage problems we were having.

This is Classic. We were having problems with players engaging to melee so they could hit every move, and then being unable to disengage. We've since realized that this is at least partly because this stage is a subtle pit, and the one platform provides a roof that effectively makes it so that if two people meet in the middle of the map, the only way for one of them to leave is exactly the way they came, which involves moving uphill.

This is Lava Platforms. It doesn't have that issue, but testers don't seem to like it. Maybe it's too dangerous, maybe it's too hard to navigate without blink. (Right now you can't jump through the platforms from beneath.)







This is Rage Cage. It's our least subtle, and most innovative, attempt to solve the disengage issue. Quite simply, there's a wall across the middle of the stage. You can't close to melee at all without Blink. This stage tested extremely well, so we may continue with this idea. I think at very least we'll ship with some 'Cage' stages if we move forward with this idea.



This is Final Destination. Named in reference to Super Smash Bros. Melee, it is the simplest possible stage, more or less, and it's amazing how much difference not having the platform makes. You can dodge attacks with jumps. You can run away, you can jump over the other player, you can reposition yourself in ways that aren't possible at all on Classic. Right now I'm pretty sure Classic is just terrible.


This is Giant Temple. It is more or less the platform layout (and loosely scale) of Hyrule Temple, again from Super Smash Bros. Melee. I wanted to put in a big stage with room to explore and fight in different environments, but I didn't want to spend too long building it (I am not a level designer, nor did I have one on hand) so I swiped an existing map layout. It plays surprisingly well, and is the only stage where all six attacks feel equally powerful to me. Blinking downward through platforms for an un-chaseable escape is a lot of fun. This stage is roughly four times as large as the others, in case scale is unclear.

This prototype is playable online now, go to fighting.loulessing.com to play. It requires two players and two Xbox 360 controllers plugged in. It still controls like crap, I'm going to rework movement before Tuesday.

Sunday, September 15, 2013

Nothing Ever Works on Lab Machines...

The prototypes are coming along nicely.

The one I'm working on (Fighting/Deckbuilding) looks like this:




It used to look like this:


But sometimes that made it look like this:



on testing machines. Right now I'm inclined to blame the hardware, although I don't have anything like the time to debug an issue in a throwaway prototype that only appears on machines I need to take a bus to access, so I just took out the GPU particles. Now it looks lame, but it's testable, which is a better place to be. The GPU particles aren't me, I'm using TCParticles, by Tiny Cube Studio. It's a very cool piece of tech, they're worth taking a look at if you have anything in Unity that requires particles that actually look good. ( http://tcparticles.tinycubestudio.com/ )

I'll try to stabilize the particle version if we decide to move forward with this prototype, on PC, with an art style that fits this sort of aesthetic. I don't want to spend the time chasing that bug right now, given the very real possibility that we'll go forward with Moles, or try to put this on OUYA, or choose an art style that doesn't work well with tons of shiny particles everywhere.

I still don't want to go forward with Moles, by the way. I've figured out what worries me about it so much. Fighting/Deckbuilding has two core systems (Fighting, drafting) and every other thing interacts more or less only with those two systems. Essentially, programming time scales linearly with features added to the game. Moles has half a dozen smaller systems (Voxels, Fog of War, simple AI, powerups, traps, a whole bunch of hazards that follow their own rules) that all interact with each other. None of these systems on its own is particularly intimidating, but each new feature has to interact with more or less every existing system, which means that programming time scales quadratically with features added to the game. Adding this to the fact that that design is still in a place where meaningful features are being added on a regular basis, and that it is already pushing what I think I can do in ten weeks, I really don't think Moles is feasible.

Sunday, September 8, 2013

Getting Started

So this is the start of a new project. Kind of. I've been working on prototypes with my team all summer, but officially this is the start of a new project. It's a little exciting, but mostly just intimidating and exhausting. It's going to be really interesting working on this game, solving new problems,  and so on, but I can already tell it's also going to be a very long year.

We're working on two prototypes.

One is a customizable platform fighter. It's fast, it's deep, it's probably fun. It wears its inspirations (Super Smash Bros, Magic: The Gathering, a little bit of League of Legends and DOTA) on its sleeve, and that's just fine for the kind of game we want it to be. I'm confident we have enough unique ideas that it is a game of its own, rather than just being derivative. From a programming perspective, it's pretty simple, or at least extremely well-defined. I have a playable prototype done, it took me like two days of solid work, and that means that if we pursue this project, currently without a name, I will probably be able to spend most of my time writing tools code for our designers to add new attacks. I like that. I enjoy writing tools code, and I'm pretty good at it.

The other is a game about mole people. Players tunnel around and try to fight each other with environmental hazards. They can't hurt other moles directly, but they can drop them into pits or flood their tunnels. It's a much more interesting, but much riskier design, in the sense that it isn't really directly preceded by anything in the same way, and as such we have very few reference points to base the gameplay upon. I'm not doing the prototype for it. Hopefully when I see the prototype I'll get a better idea of what this game is actually intended to be. From a programming perspective, this tentatively looks like a nightmare. It doesn't look hard, there aren't really any specific threatening challenges in the design (walk, climb, break tiles, eat worms, gain powers, drown moles) but there are a ton of assorted systems and fiddly things that can occupy and endless amount of my time. If we pursue this project, it threatens to occupy my every waking hour for the rest of the year, while never presenting a single interesting challenge. This game is called Underminers, because if nothing else this team is very good at coming up with pages and pages of puns about moles.

Obviously, my vote right now is for the first one.